A Case Report of Zollinger Ellison Syndrome and Review of the Literature

Hiroyuki KATO, Eiji SHIMOZAWA, Tetsufumi KOJIMA and Tatsuzo TANABE

ABSTRACT: There is much controversy concerning the mode of therapy for patients in whom Zollinger-Ellison syndrome is strongly suspected but a tumor can not be located. We recently experienced a patient with Zollinger- Ellison syndrome presenting with melena in whom an attempt to stop the bleeding by H-2 antagonists failed and an emergency operation had to be carried out. At laparotomy, no tumor was found in the pancreas, duodenum or stomach wall and there was no specific swelling in any of the lymph nodes. A total gastrectomy was thus done with lymphadenectomy and a histopath'ological examination revealed two gastrinomas in the lymph

,/ nodes of the gastrmoma triangle. Postoperative secretin tests with 2 u /kg of secretin have been negative even 6.5 years later, and the patient is now well and working as a full time teacher. In this case, an emergency total gastrectomy was performed for uncontrolled bleedingl but we want to stress the importance of lymphadenectomy based on the findings of the frozen section and changes in gastrin levels.

KEY WORDS: Zollinger-Ellison syndrome, gastrinoma, lymphadenec- ,tomy, total gastrectomy

INTRODUCTION

The diagnosis of Zollinger-Ellison syn- drome has become relatively easy with the extablishment of the gastrin immunoreactive method, especially since the gastrin stimula- tion method was introduced. Localization of the gastrinoma is usually done preopera- tively by angiography, CT scan and selective portal venous sampling, however, in some cases, pre-operative or intraoperative locali- zation of the tumor is difficult. Moreover, there is much controversy concerning the

The Second Department of Surgery, Hokkaido Uni- versity School of Medicine, Sapporo, Hokkaido, Japan

Reprint requests to: Hiroyuki Kato, MD, The Second Department of Surgery, Hokkaido University School of Medicine, 060 N 14 W5 Kita-ku, Sapporo, Hokkaido 060, Japan

recommended mode of therapy for such patients. 1-4 The following is a report on the mode of therapy chosen by us for such a patient, together with a discussion of the literature on this subject.

CASE REPORT

A 44 year old male was admitted to our hospital in January, 1981 with epigastralgia and melena. On admission the serum gastrin level was found to be 130-695 pg /ml and rose to 1930 pg/ml following a secretin (2 U/kg) stimulation. Gastroduodenoscopy re- vealed a deep ulcer in the second portion of the duodenum and he was diagnosed as Zollinger-Ellison syndrome. Treatment with cimetidine (600 mg/day) was commenced and an attempt made to localize the tumor by t h e following examinations. CT scan of the

JAPANESE JOURNAL OF SURGERY, VOL. 21, NO. 1 pp. 105-109, 1991

106 Katoh et al.

abdomen and visceral angiography showed a normal pancreas and liver while selective portal venous Sampling demonstrated no step up of serum IRG gradient in any part of the splenic, superior mesenteric or portal veins. During the period whilst the examina- tions were being carried out, the melena pers is ted despi te i nc reased amounts o f cimetidine (1200 mg/day).

Modification to ranitidine (200 mg/day) had no positive effect, and the melena con- tinued. During the 4th month after his ad- mission the bleeding increased, necessitating a blood transfusion of 6000 ml over a period of 5 days which was followed by an emer- gency laparotomy, performed in May, 1981. The liver surface was normal but an ulcer was palpated in the second portion of the duodenum and thus, under direct visualiza- tion using bimanual palpation, the entire pancreas and duodenum were carefuily ex- amined but no tumor was able to be found in the gastrinoma triangle. Pyloroduodenotomy was performed and the duodenal wall was examined but, except for a bleeding ulcer, no tumor was found and thus, a total gastrec- tomy was performed, together with lymph- adenectomy.

Histopathological examination by Hema- toxylin-eosin staining and Grimelius silver staining showed evidence of a neuroendo- crine tumor in only 2 lymph nodes adjacent to the head o f the pancreas (Fig. 1, 2). The existence of a gastrin Producing tumor was subsequent ly conf i rmed by an immuno- histological examination, using the fluores- cent contrast method (Fig. 3). The antisera prepared for this study were rabbit antigastin serum, donated by Dr. T. Yabana, Sapporo Medical college, Sapporo, Japan . Large amounts of gastrin secreting granules, com- posed o f e lec t ron-dense and half-empty granules, were also verified by e lec t ron microscopy (Fig. 4). A serial sectioning of 5 mm intervals of the resected stomach and duodenum was performed, but no tumor found.

The patient's post-operative course was

Jpn. J. Surg. January 1991

Fig. 1. Photomicrograph of a lymph node containing metastatic gastrinoma cells. (HE X100)

Fig. 2. Photomicrograph of the tumor by Grimelius silver staining. The argyrophil reaction is positive. (•

Fig. 3. Photomicrography of the tumor by the fluorescent contrast method. Gastrin producing cells were detected. A rabbit antigastrin serum was used. (X400)

Volume 21 Number 1 A case of ZE

Fig. 4. Electron micrography of the tu- mor. Large amounts of gastrin secreting granules composed of electron-dense and half-empty granules can be seen. (•

uneventful and subsequent basal and secre- tin (2 U/kg) stimulated gastrin levels have been normal even 8 years later. He has been working as a teacher of pharmacology for seven years since his operation, and his body weight is now 3 kg more than before the operat ion.

DISCUSSION

The principle mode of therapy for Zol- linger-Ellison syndrome is resection of the primary tumor even though H-2 receptor antagonists have become more generally used today? -1~ In some patients, however, the pr imary tumor can not be located by angiography, CT scan or selective venous sampling and there is much controvery con- cerning the r ecommended mode of therapy for such patients? -6

Some r e c o m m e n d long- term follow-up' together with H-2 receptor antagonist adminl istration, 8,9 while others r ecommend total resection of the stomach. Bornman et al., 3 who treated 5 patients in whom a tumor was unable to b e localized preoperatively, by simple excision or ','shelling out" of the" macroscopic tumor, achieved good results, and thus claim that the place of total gas-

syndrome 107

trectomy in the treatment o f such patients is questionable. In 2 of these 5 patients, how- ever, H-2 receptor antagonist administration was later required. In 9 of 27 patient de- scribed by T h o m p s o n et ai., ~ tumor localiza- tion was unclear, yet no operative death occurred following either total or subtotal gastrectomy and long-term survival together with excellent nutritional results were re- ported in all patients.

Stabile et al? named the area surrounding the pancreatic head the "gastr inoma trian- gle" and emphasized the necessity to care- fully examine that area and remove its lymph nodes and tissue no matter whether a pri- mary tumor is found there or not. T h e reason for this is that occult tumors are often found in that area. Thompson et al. 4 de- scribed two patients in whom metastasis was considered even though both had normal gastrin levels after excision of a lymph node containing tumor, emphasizing the possi- bility that the gastr inoma had arisen f rom the lymph nodes. In his recent reports, however, he emphasized that occult tumors were most often found in the wall of the duodenum and might be as small as 2 mm in diameter . He stressed that previously re- ported patients with primary "lymph node gas t r inomas" p robab ly had had p r imary tumors excised at a previous gastric opera- tion which had been overlooked. H Stabile et al. 2 also described two patients of gastr inoma arising in the lymph nodes which were cured following the discovery of occult tumors in the anatomic triangle. Among the 52 patients he studied, a tumor could not be found in 10 patients, but the 10 year survival rate of those patients was 90 per cent. In the cases of gas- t r inoma arising in lymph nodes, very small occult tumors can exist in the s tomach or duodenum which are easy to be over looked even in a careful histological examination. Delcore et al. TM described that gastrinomas found in lymph nodes might be metastatic f rom either submucosal duodenal primaries or aberrant primary nodal lesions and that in the absence of hepatic involvement, lymph

108

node gastrinomas, whether pr imary o r meta- static had a good prognosis and should no t deter aggressive surgical treatment. Zoll inger recent ly repor ted that his original pat ient in 1955 i n which metastases were found in the lymph nodes, has shown gradually increas- ing gastrin levels over the subsequent 30 yearsY, ~3

F rom the above m e n t i o n e d literature, it appears that current ly there are no dec ided principles o f therapy for such patients. Dur- ing the last 10 years, however, a gastrin immunoreac t ive m e a su re m e n t me thod has ben established, and it is now possible to. pe r fo rm these measurements within 30 to 60 min. 14a5 Taking this into account, the pat ient in ques t ion can be reviewed. In this patient, H-2 r e c e p t o r a n t a g o n i s t was ineffect ive, necess i t a t ing an e m e r g e n c y o p e r a t i o n to achieve definite hemostasis. No tumor was f o u n d dur ing the laparotomy and no en- larged lymph glands were detected in the "gas t r inoma triangle". Total gastrectomy was thus pe r fo rmed and yet m a n y quest ions c o n c e r n i n g the correct j u d g m e n t o f this decision remain unsolved. O n e possibility was to pe r fo rm an ex tended lymphadenec- tomy, even t hough it was an emergency case o f bleeding, and then decide accord ing to the results o f the f rozen spec imen a n d intraoperat ive secretin test using an intra- venous injection o f 2 U / k g o f secretin. TM A rapid rad io immunoassay o f serum IRG can be pe r fo rmed by shor ten ing the incuba t ion time with ant ibodies as repor ted by Azuma a n d Imamura . 15 I f a tumor ha d b e e n f o u n d in the extracted lymph nodes and the intra- operative secretin test had b e e n negative, then the s tomach should not have b e e n resected, however, if the intraoperative secre- tin test had b e e n positive, we could have conc luded that the tumor f o u n d in the lymph nodes was a metastafivc tumor, In such pa- tients, duodenos tom y and intraluminal ex- p lorat ion should be considered, enab l ing total gastrectomy to be avoided? H s If, on the o the r hand, no tumor had b e e n found in any o f the extracted lymph nodes or duodena l

Jpn. J. Surg. Katoh et al. January 1991

wall, total gastrectomy would have b e e n unavoidable.

I n our patient, a total gas t rec tomy was per- fo rmed primarily to stop the b leed ing f rom the duodena l ulcer, accompan ied by lymph- adenectomy. However , for patients in w h o m lymphadenec tomy and a detailed examina- t ion o f the d u o d e n u m together with an intraoperative secretin test can first be per- formed, the n e e d for total gastrectomy can be restricted to those cases in which, it would be unavoidable.

(Received for publicat ion o n Aug. 21, 1989)

REFERENCES

1. Stabile BE, Morrow DJ, Passaro E Jr. The'. gas- trinoma triangle: Operative implications. Am J Surg 1984; 147: 25-31.

2. Stabile BE, Passaro E Jr. Benign and malignant gastrinoma. AmJ Surg 1985; 149: 144-150.

3. Bornman PC, Marks IN, Mee AS, Price S. Favour- able response to conservative surgery for extra- pancreatic gastrinoma with lymph node meta- stases. BrJ Surg 1987; 74: 198-201.

4. Thompson NW, Vinik AI, Eckhauser FE, Strodel WE. Extrapancreatic gastrinoma. Surgery 1985; 98: 1113-1120.

5. ThompsonJC, Lewis BG, Wiener I, Townsend CM Jr. The role of surgery in the Zollinger-Ellison syndrome. Ann Surg t983; 197: 594-607.

6. Bonfils S, LandorJH, Mignon M, Hervoir P. Reults of surgical management in 92 consecutive patients with Zollinger-Ellison syndrome. Ann Surg 1981; 194: 692-695.

7. Zollinger RM. Gastrinoma: Factors influencing prognosis. Surgery 1985; 97: 49-54.

8. Richardson CT, Peters MN, Feldman M, McClelland RN, Walsh JH, Cooper KA, Willeford G, Dicker- man RM, Fordtran JS. Treatment of Zollinger- Ellison syndrome with exploratory laparotomy, proximal gastric vagotomy, and H2-receptor antagonists. Gastroenterology 1985; 89: 357-367.

9. MalageladaJR, Edis AJ, Adson MA, HeerdenJA, Go VLW. Medical and surgical in the management of patients with gastrinoma. Gastroenterology 1983; 84: 1524-i532.

10. Deveney CW, Deveney KS, Way LW. The Zollinger- Ellison syndrome--23 years later. Ann Surg 1978; 188: 384-393.

11. Thompson NW, Vinik AI, Eckhauser FE. Micro- gastrinomas of the duodenum: A cause of failed operations for the Zollinger-Ellison syndrome.

Volume 21 Number 1

A case of ZE syndrome 109

Ann Surg 1989; 209: 396-404. 12. Delcore R Jr, Cheung LY, Friesen SR. Outcome of

lymph node involvement in patients with the Zollinger-Ellison syndrome. Ann Surg 1988; 208: 291-298.

13. Zollinger RM. Treatment of gastrinoma. Mount Sinai Med 1984; 51: 401-403.

14. Imamura M, Takahashi K, Isobe Y, Hattofi Y, Satomura K, Tobe T. Curative resection of multiple

gastrinomas aided by selective arterial secretin injection test and intraoperative secretin test. Ann Surg 1989; 210: 710-718.

15. Azuma T, Imamura M, Shimada Y, Inokuchi H, Kawai E. Intraoperative secretin test for the rapid evaluation of curative operation in a case of Zollinger-Ellison syndrome. J Gastroent Hepat 1988; 3: 111-115.